BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Kuwait Foundation Lectures
SUMMARY:Estimation of Large Covariance Matrix - Professor
Jianqing Fan (Princeton)
DTSTART;TZID=Europe/London:20080115T170000
DTEND;TZID=Europe/London:20080115T180000
UID:TALK8783AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/8783
DESCRIPTION:Large dimensionality comparable to the sample size
is a common feature as in portfolio allocation\,
risk management\, genetic network and climatology.
In this talk\, we first use a multi-factor model
to reduce the dimensionality and to estimate the
covariance matrix for portfolio allocation and ris
k management. The impacts of dimensionality on th
e estimation of covariance matrix and its inverse
are examined. We identify the situations under wh
ich the factor approach can gain substantially the
performance and the cases where the gains are onl
y marginal\, in comparison with the sample covaria
nce matrix. Furthermore\, the impacts of the covar
iance matrix estimation on portfolio allocation an
d risk management are studied. In other class of
problems such as genetic network or climatology\,
sparsity of the covariance matrix or its inverse a
rises natural. We then estimate the large covaria
nce matrix estimation by exploiting its sparsity u
sing the penalized likelihood method. Sampling pr
operty is established and new algorithms are propo
sed.\n\n
LOCATION:Wolfson Room (MR 2) Centre for Mathematical Scienc
es\, Wilberforce Road\, Cambridge
CONTACT:Helen Innes
END:VEVENT
END:VCALENDAR